Methods and Apparatus to Generate Corrected Online Audience Measurement Data

1. A method comprising: obtaining, by an audience measurement computing system, census data for a first episode of a video program, the census data indicative of streaming-media impressions for the first episode; obtaining, at an impression collection server, a first beacon request indicating that a computing device accessed a second episode of the video program via the Internet; based on receiving the first beacon request, instructing the computing device to send a second beacon request to a database proprietor, wherein the database proprietor determines demographic data corresponding to a user associated with the computing device based on the second beacon request; obtaining, by the audience measurement computing system from a server of the database proprietor, aggregated demographic data for a subset of episodes of the video program, wherein the subset of episodes comprises the second episode of the video program and at least one additional episode of the video program, and wherein the aggregated demographic data for each respective episode of the subset of episodes is indicative of a demographic distribution of streaming-media impressions for the respective episode across a plurality of demographic groups; in response to determining that the database proprietor does not have sufficient aggregated demographic data for the first episode, determining a coverage adjustment vector based on survey calibration data including a total number of respondents in a demographic group of the plurality of demographic groups; based on the aggregated demographic data of the subset of episodes, calculating, by the audience measurement computing system, respectively for each demographic group of the plurality of demographic groups, video program-level impressions data for the video program by summing impressions of demographic groups across the subset of episodes of the video program to determine corresponding video-program level impressions summed across the subset of episodes of the video program; determining an impression demographic distribution for the subset of episodes by dividing the video-program level impressions for each demographic group by a sum of the video-program level impressions across all demographic groups of the plurality of demographic groups; using, by the audience measurement computing system, the impression demographic distribution for the subset of episodes as a basis to distribute the streaming-media impressions for the first episode to respective demographic groups of the plurality of demographic groups; and generating, by the audience measurement computing system, ratings information for the first episode using the distributed streaming-media impressions for the first episode.

2. The method of claim 1, wherein using the impression demographic distribution for the subset of episodes as the basis to distribute the streaming-media impressions for the first episode comprises: determining a percentage of streaming-media impressions for the subset of episodes that are associated with a demographic group of the plurality of demographic groups; and using the percentage as a basis to impute a portion of the streaming-media impressions for the first episode to the demographic group.

3. The method of claim 1, wherein the at least one additional episode of the video program comprises a third episode, the method further comprising: determining a combined demographic distribution for the video program using the aggregated demographic data for the second episode and the aggregated demographic data for the third episode, and wherein using the impression demographic distribution for the subset of episodes as the basis to distribute the streaming-media impressions for the first episode comprises using the combined demographic distribution for the video program as the basis to distribute the streaming-media impressions for the first episode.

4. The method of claim 1, wherein obtaining the census data for the first episode comprises receiving, at a server of the audience measurement computing system, a network communication comprising data indicating that a second computing device accessed the first episode via the Internet, the network communication triggered by beacon instructions associated with the first episode.

5. The method of claim 4, wherein the database proprietor determines the aggregated demographic data for the second episode using demographic data known by the database proprietor for users that accessed the second episode via the Internet.

6. The method of claim 1, wherein the first episode or an application used to access the first episode is tagged with beacon instructions that are downloaded to computing devices when the computing devices access the first episode, wherein the beacon instructions cause the computing devices to transmit beacon requests to the impression collection server to facilitate the impression collection server collecting the streaming-media impressions.

7. The method of claim 1, wherein the coverage adjustment vector is a ratio of (i) a total number of respondents in the demographic group with access to a device type associated with the video program to (ii) a number of respondents recognized by the database proprietor for the demographic group.

8. A computing system comprising a processor and a memory, the computing system configured to perform a set of acts comprising: obtaining census data for a first episode of a video program, the census data indicative of streaming-media impressions for the first episode; obtaining, at an impression collection server, a first beacon request indicating that a computing device accessed a second episode of the video program via the Internet; based on receiving the first beacon request, instructing the computing device to send a second beacon request to a database proprietor, wherein the database proprietor determines demographic data corresponding to a user associated with the computing device based on the second beacon request; obtaining, from a server of the database proprietor, aggregated demographic data for a subset of episodes of the video program, wherein the subset of episodes comprises the second episode of the video program and at least one additional episode of the video program, and wherein the aggregated demographic data for each respective episode of the subset of episodes is indicative of a demographic distribution of streaming-media impressions for the respective episode across a plurality of demographic groups; in response to determining that the database proprietor does not have sufficient aggregated demographic data for the first episode, determining a coverage adjustment vector based on survey calibration data including a total number of respondents in a demographic group of the plurality of demographic groups; based on the aggregated demographic data of the subset of episodes, calculating, respectively for each demographic group of the plurality of demographic groups, video program-level impressions data for the video program by summing impressions of the plurality of demographic groups across the subset of episodes of the video program to determine corresponding video-program level impressions summed across the subset of episodes of the video program; determining an impression demographic distribution for the subset of episodes by dividing the video-program level impressions for each demographic group by a sum of the video-program level impressions across all demographic groups of the plurality of demographic groups; using the impression demographic distribution for the subset of episodes as a basis to distribute the streaming-media impressions for the first episode to respective demographic groups of the plurality of demographic groups; and generating ratings information for the first episode using the distributed streaming-media impressions for the first episode.

9. The computing system of claim 8, wherein using the impression demographic distribution for the subset of episodes as the basis to distribute the streaming-media impressions for the first episode comprises: determining a percentage of streaming-media impressions for the subset of episodes that are associated with a demographic group of the plurality of demographic groups; and using the percentage as a basis to impute a portion of the streaming-media impressions for the first episode to the demographic group.

10. The computing system of claim 8, wherein the at least one additional episode of the video program comprises a third episode, and the set of acts further comprises: determining a combined demographic distribution for the video program using the aggregated demographic data for the second episode and the aggregated demographic data for the third episode, and using the impression demographic distribution for the subset of episodes as the basis to distribute the streaming-media impressions for the first episode comprises using the combined demographic distribution for the video program as the basis to distribute the streaming-media impressions for the first episode.

11. The computing system of claim 8, wherein obtaining the census data for the first episode comprises receiving, at a server of the computing system, a network communication comprising data indicating that a second computing device accessed the first episode via the Internet, the network communication triggered by beacon instructions associated with the first episode.

12. The computing system of claim 11, wherein the database proprietor determines the aggregated demographic data for the second episode using demographic data known by the database proprietor for users that accessed the second episode via the Internet.

13. The computing system of claim 8, wherein the first episode or an application used to access the first episode is tagged with beacon instructions that are downloaded to computing devices when the computing devices access the first episode, wherein the beacon instructions cause the computing devices to transmit beacon requests to the impression collection server to facilitate the impression collection server collecting the streaming-media impressions.

14. The computing system of claim 8, wherein the coverage adjustment vector is a ratio of (i) a total number of respondents in the demographic group with access to a device type associated with the video program to (ii) a number of respondents recognized by the database proprietor for the demographic group.

15. A non-transitory computer-readable medium having stored therein instructions that when executed by a computing system cause the computing system to perform a set of acts comprising: obtaining census data for a first episode of a video program, the census data indicative of streaming-media impressions for the first episode; obtaining, at an impression collection server, a first beacon request indicating that a computing device accessed a second episode of the video program via the Internet; based on receiving the first beacon request, instructing the computing device to send a second beacon request to a database proprietor, wherein the database proprietor determines demographic data corresponding to a user associated with the computing device based on the second beacon request; obtaining, from a server of the database proprietor, aggregated demographic data for a subset of episodes of the video program, wherein the subset of episodes comprises the second episode of the video program and at least one additional episode of the video program, and wherein the aggregated demographic data for each respective episode of the subset of episodes is indicative of a demographic distribution of streaming-media impressions for the respective episode across a plurality of demographic groups; in response to determining that the database proprietor does not have sufficient aggregated demographic data for the first episode, determining a coverage adjustment vector based on survey calibration data including a total number of respondents in a demographic group of the plurality of demographic groups; based on the aggregated demographic data of the subset of episodes, calculating, respectively for each demographic group of the plurality of demographic groups, video program-level impressions data for the video program by summing impressions of the plurality of demographic groups across the subset of episodes of the video program to determine corresponding video-program level impressions summed across the subset of episodes of the video program; determining an impression demographic distribution for the subset of episodes by dividing the video-program level impressions for each demographic group by a sum of the video-program level impressions across all demographic groups of the plurality of demographic groups; using the impression demographic distribution for the subset of episodes as a basis to distribute the streaming-media impressions for the first episode to respective demographic groups of the plurality of demographic groups; and generating ratings information for the first episode using the distributed streaming-media impressions for the first episode.

16. The non-transitory computer-readable medium of claim 15, wherein using the impression demographic distribution for the subset of episodes as the basis to distribute the streaming-media impressions for the first episode comprises: determining a percentage of impressions for the subset of episodes that are associated with a demographic group of the plurality of demographic groups; and using the percentage as a basis to impute a portion of the streaming-media impressions to the demographic group.

17. The non-transitory computer-readable medium of claim 15, wherein the at least one additional episode of the video program comprises a third episode, and the set of acts further comprises: determining a combined demographic distribution for the video program using the aggregated demographic data for the second episode and the aggregated demographic data for the third episode, and using the impression demographic distribution for the subset of episodes as the basis to distribute the streaming-media impressions for the first episode comprises using the combined demographic distribution for the video program as the basis to distribute the streaming-media impressions for the first episode.

18. The non-transitory computer-readable medium of claim 15, wherein obtaining the census data for the first episode comprises receiving, at a server of the computing system, a network communication comprising data indicating that a second computing device accessed the first episode via the Internet, the network communication triggered by beacon instructions associated with the first episode.

19. The non-transitory computer-readable medium of claim 18, wherein the database proprietor determines the aggregated demographic data for the second episode using demographic data known by the database proprietor for users that accessed the second episode via the Internet.

20. The non-transitory computer-readable medium of claim 15, wherein the first episode or an application used to access the first episode is tagged with beacon instructions that are downloaded to computing devices when the computing devices access the first episode, wherein the beacon instructions cause the computing devices to transmit beacon requests to the impression collection server to facilitate the impression collection server collecting the streaming-media impressions.